| --- |
| tags: |
| - generated_from_trainer |
| datasets: |
| - roneneldan/TinyStories |
| metrics: |
| - accuracy |
| model-index: |
| - name: gpt2_m100_tiny-stories_1024_dpos |
| results: |
| - task: |
| name: Causal Language Modeling |
| type: text-generation |
| dataset: |
| name: roneneldan/TinyStories |
| type: roneneldan/TinyStories |
| metrics: |
| - name: Accuracy |
| type: accuracy |
| value: 0.6900624734182258 |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/scads-nlp/morph-gpt_gpt2_tiny-stories_dpos/runs/y12q0b1d) |
| # gpt2_m100_tiny-stories_1024_dpos |
|
|
| This model is a fine-tuned version of [](https://huggingface.co/) on the roneneldan/TinyStories dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 1.1579 |
| - Accuracy: 0.6901 |
|
|
| ## Model description |
|
|
| More information needed |
|
|
| ## Intended uses & limitations |
|
|
| More information needed |
|
|
| ## Training and evaluation data |
|
|
| More information needed |
|
|
| ## Training procedure |
|
|
| ### Training hyperparameters |
|
|
| The following hyperparameters were used during training: |
| - learning_rate: 5e-05 |
| - train_batch_size: 32 |
| - eval_batch_size: 32 |
| - seed: 42 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: linear |
| - num_epochs: 1.0 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | Accuracy | |
| |:-------------:|:------:|:-----:|:---------------:|:--------:| |
| | 2.801 | 0.0506 | 1000 | 2.3554 | 0.4658 | |
| | 1.8977 | 0.1012 | 2000 | 1.7248 | 0.5834 | |
| | 1.658 | 0.1518 | 3000 | 1.5498 | 0.6145 | |
| | 1.5426 | 0.2024 | 4000 | 1.4542 | 0.6321 | |
| | 1.4721 | 0.2530 | 5000 | 1.3930 | 0.6435 | |
| | 1.4237 | 0.3036 | 6000 | 1.3497 | 0.6517 | |
| | 1.387 | 0.3543 | 7000 | 1.3162 | 0.6580 | |
| | 1.3537 | 0.4049 | 8000 | 1.2899 | 0.6633 | |
| | 1.3306 | 0.4555 | 9000 | 1.2683 | 0.6676 | |
| | 1.3127 | 0.5061 | 10000 | 1.2474 | 0.6716 | |
| | 1.2925 | 0.5567 | 11000 | 1.2326 | 0.6745 | |
| | 1.2779 | 0.6073 | 12000 | 1.2171 | 0.6778 | |
| | 1.262 | 0.6579 | 13000 | 1.2051 | 0.6802 | |
| | 1.2502 | 0.7085 | 14000 | 1.1949 | 0.6823 | |
| | 1.2413 | 0.7591 | 15000 | 1.1852 | 0.6843 | |
| | 1.2354 | 0.8097 | 16000 | 1.1773 | 0.6857 | |
| | 1.2254 | 0.8603 | 17000 | 1.1699 | 0.6874 | |
| | 1.2186 | 0.9109 | 18000 | 1.1639 | 0.6887 | |
| | 1.2155 | 0.9615 | 19000 | 1.1597 | 0.6897 | |
|
|
|
|
| ### Framework versions |
|
|
| - Transformers 4.42.3 |
| - Pytorch 2.2.2+cu121 |
| - Datasets 2.20.0 |
| - Tokenizers 0.19.1 |
|
|
